EP0059745A1

EP0059745A1 - Method and device for the localisation and analysis of sound emissions.

Info

Publication number: EP0059745A1
Application number: EP81902643A
Authority: EP
Inventors: Horst-Artur Crostack
Original assignee: GEWERTEC GESELLSCHAFT fur WERKSTOFFTECHNIK MBH; GEWERTEC GmbH
Current assignee: GEWERTEC GESELLSCHAFT fur WERKSTOFFTECHNIK MBH; GEWERTEC GmbH
Priority date: 1980-09-10
Filing date: 1981-09-05
Publication date: 1982-09-15
Also published as: US4459851A; WO1982000893A1; JPS57501448A; JPH0415415B2; EP0059745B1

Abstract

Il s'agit d'un procede de localisation et d'analyse d'emissions sonores, dans lequel la position de la source d'emission des sons est determinee au moyen de l'arrivee d'une impulsion sonore en differents endroits de mesure. A cet effet, on mesure en trois ou plusieurs parties voisines de chaque endroit de mesure, dont la largeur est plus petite que le diametre du champ sonore attendu, l'instant d'arrivee d'une impulsion ou d'une caracteristique d'impulsion. Ensuite, on determine la difference des temps de propagation pour les parties de chaque endroit de mesure et, a partir des valeurs de ces differences des temps de propagation pour chaque endroit de mesure, on determine la direction incidente du son. Au moyen des valeurs definissant ces directions et des coordonnees des positions des endroits de mesure, on determine alors la position de la source d'emission sonore. Le dispositif pour la mise en oeuvre de ce procede comprend, pour chaque tete de mesure (E) de recepteur, trois elements convertisseurs (1) distincts ou davantage, dont la largeur totale est plus petite que le diametre du champ sonore attendu.It is a method of locating and analyzing sound emissions, in which the position of the source of sound emission is determined by means of the arrival of a sound pulse at different measurement locations. For this purpose, in three or more parts adjacent to each measurement location, the width of which is smaller than the diameter of the expected sound field, the instant of arrival of a pulse or of a pulse characteristic is measured. . Next, the difference in the propagation times for the parts of each measurement location is determined and, from the values of these differences in the propagation times for each measurement location, the incident direction of the sound is determined. By means of the values defining these directions and the coordinates of the positions of the measuring points, the position of the source of sound emission is then determined. The device for implementing this method comprises, for each receiver measuring head (E), three or more distinct converter elements (1), the total width of which is smaller than the diameter of the expected sound field.

Description

Method and device for sound emission location and analysis

The invention relates to a method for locating and analyzing sound emissions, in which the position of the sound emission source is determined from the arrival of a sound pulse at various measuring points. Furthermore, the invention relates to a device for sound emission location and analysis which works according to this method and which is intended in particular for the testing of devices and components for damage sites and for monitoring devices and components.

The problem underlying the invention is explained below with reference to the main field of application of the invention. However, the invention is not only applicable in this field, but in any other field in which the source of the sound emissions is to be located from the arrival of a sound pulse;

When operating machines and systems, noises can occur that arise inside the materials and are related to the beginning of failure. For example, emerging cracks in the micro range emit such sound waves, sometimes at times well before the failure. These sound waves are used for early damage detection, whereby the degree of damage and the progress of the failure can be recorded. In addition to the analysis of the processes - definition of the type of sound sources - in particular The localization of the sound source is of great importance for remedial measures.

As is known, the location of the sound sources is carried out in such a way that the running time differences are measured which occur on the path of the sound from the sound source to different receivers which are arranged individually and at greater distances from one another around the source, and from these running time differences the position of the sound source is calculated based on the position coordinates of the receiver. When applying this procedure to existing or emerging damage spots, several receivers are attached to the surface of the test object at intervals of 1 to 5 m. If a sound impulse hits the receivers, the signals they emit trigger a runtime count. The running times of the next neighboring probes are compared and the location of the sound source is determined from the running differences and the receiver coordinates. (H. Bretfeld, A. Möller and H.-A. Crostack: Application of sound emission analysis when a pressure vessel is loaded with a pulsating internal pressure; Materialprüfung 19 (1977) 11, pp. 467/70).

For this known determination of the location to lead to correct results, the following essential requirements must be met:

1. The speed, which the Sehall wave develops on its way γ from the sound source to the receiver, must be known, since otherwise the running time does not count that of the sound wave traveled distance and thus the distance of the sound source from the receiver can be calculated.

2. The sound impulses must not occur in close chronological order, but must have clear distances. If another pulse occurs in other places before. the wave of the first pulse has reached all the receivers required for the location, there is a faulty location. (H.-D. Steffens and H.-A. Cros'tack: Influencing factors in the analysis of sound emissions; Zeitschrift für Werkstofftechnik (1973) 8, p. 442/7).

3. The sound impulses have to be very similar, since a change in the wave type or the pulse form (eg the amplitude or the rise time) can trigger the electronic recording devices (triggers) at different times (HD Steffens, D. Krempel, D. Stegemann and H.-A. Cro stack: Analysis of signals of sound emission; Zeitschrift für Werkstofftechnik 6 (L975) 3, p. 88/94.

So that "critical" sources are located and not those of background noise, the. Sound source can be identified. For this purpose, an analysis is required which, in the known method and the devices operating according to it, can only be carried out at the same time as the location, and not simultaneously with it.

In the case of the sound pulses to be located which originate from sound emission events, none of the conditions listed above can be met, as will be explained below with reference to the numbers used above. Regarding number 1: Speed of sound: When a single sound emission event occurs, which usually occurs in a pulsed manner, longitudinal and transverse waves are released, which also have an angle-dependent intensity (directional beam characteristic). As a result, these waves have different intensities depending on the event and they propagate at different speeds.

When passing through a solid (system, machine), these waves are reflected, damped and converted. In addition to the pure fashion conversion, guided waves (surface, plate, tube, rod waves, etc.) can also occur, some of which have very high dispersion. However, this means that the speed of propagation depends on unknown values, namely on the frequency spectrum. In addition, since this conversion - depending on the geometry of the body - depends on the travel path, the receiver has a different average speed for each distance of the sound source.

In the known locating methods and the devices operating according to them, the speed is specified as constant, deviating from the actual circumstances. This results in significant deviations of the result from the real conditions.

The same applies if the sound passes through different media at different speeds, such as a water-filled container. Here, too, the specification of a constant speed leads to errors in the evaluation of the measurement results use of the sound emission method for test purposes has so far not seemed to make sense.

To number 2: Time intervals of the sound pulses: The condition that sound pulses occur far apart in time can only be found in individual cases - e.g. in a pressure test with controllable load conditions. When events occur in close succession, e.g. plasti rather deformation, phase changes in the structure or manufacturing noises, such as occur in welding and soldering processes, has so far been impossible to locate.

Regarding number 3: Similarity of the impulses:

This third prerequisite cannot be met in real structures (H.-D. Steffens, D. Stegemann, D. Krappe! And H.-A. Crostack: Analysis of signals from sound emissions; Zeitschrift für Werk-Stofftechnik 6 (1975 ) 3, pp. 88/94). The sound source emits longi tudinal and transverse waves, the intensity of which varies depending on the direction. As a result of the propagation with the wave conversion mentioned under item 1 and the formation of guided waves, the intensity relationships to one another continue to change. Since the spectra also change due to the attenuation and the receivers are designed for specific resonance frequencies, this can occur that the electronics of one receiver respond to the longitudinal wave components of the pulse, and that of another to the transverse components (or surface waves, for example). This causes large errors in the location and the assignment of the sound source is not possible. For these reasons, the determination of the location of sound sources by means of the known method already has great difficulties in simple components, such as plates or pipes. With complicated geometries, such as those found in machines and systems, the difficulties become so great that location is impossible.

The invention has for its object to provide a Verfaljren and a device with which it is possible in a simple manner even with unknown speed of propagation of the pulse, with a high temporal signal sequence, with heavily noisy signals and with greatly changed pulses with a longer path as well as complicated component geometries to locate (place of origin), and thus significantly expand the field of application of the noise emission method when checking for damage.

The invention consists in that in three or more closely adjacent sections of each measuring point, the width of which is smaller than the blade diameter to be expected, the moment of impulse measured, from this directly or after a real-time analysis, the transit time difference for the sections of each measuring point determined and from these transit time difference values the direction in which the sound emission source 1 e lies is determined for each measuring point, after which the position of the sound emission source is determined from the direction values and the position coordinates of the measuring points.

Moreover, the invention consists in a device to practice this method, in which each receiver probe has three or more separate transducer elements, the overall width of which is less than the expected sound field diameter.

Further features of the invention are the subject of the dependent claims.

An example embodiment of a device for carrying out the method according to the invention is described below with reference to the block diagram of such a device shown in the drawing. For the sake of simplicity, only one of the receiver test heads of the device is shown and described, since the test heads correspond in their basic structure and mode of operation.

In the embodiment shown, the receiver E is composed of three individual sound transducer elements 1. However, several transducer elements 1 can also be assembled to form a test head E, provided that their overall width does not exceed the maximum limit provided. The diameter of the receiver E is small compared to the extent of the sound field to be expected, i.e. e.g. smaller than 25 mm so that the elements l lie within a sound field range of the source, e.g. with zigzag waves within the 6 dB drop. The elements l can be made of any

Transducer materials and structures (piezoelectric, electrodynamic, acousto-optical, magnetostrictive, etc.) exist. Each converter element 1 is - in the order listed - an amplifier and impedance converter 2, a digitizer 3, a delay 4, a counter 5 and a clock controller 8. The digitizers 3 are also connected on the output side to a counter starter 9, which serves for the simultaneous switching on of all counters 5; it is switched on by the first incoming digitized pulse, so that all counters 5 are started by this. The delay time of the delays 4 is the same for all channels of a receiver E, for example 20ns. The counters 5 are stopped by the first pulse arriving with the predetermined delay via the receiver channel assigned to it. A timer 7 is used to measure the absolute time when the counters 5 start. Like the counters 5, it is connected with its signal output to the inputs of a microprocessor into which, controlled by the clock controller 8, the measured values - the counters 5 and of the timer 7 can be entered. For each receiver E there is a separate computer 6 of this type, which determines the transit time differences between the individual elements 1 of a receiver from the transit time values entered and calculates the direction from which the pulse has arrived, and in parallel as a parameter for the pulse saves absolute time at counter start and then - with a delay - clears counter 5.

All microprocessors 6 of a device are connected with their outputs to a computer 12 of the device. In the drawing are 10 and 11 Refers to lines via which the characteristic values of other receivers are entered into the computer 12. The output unit 13 is connected downstream of the computer 12. It is possible to connect filters (not shown) or analyzes (not shown) between the impedance converters 2 and the digitizers 3, which filters can extract its characteristic from a noisy signal.

The device works as follows:

When a sound event occurs, the source of which is to be located, the sound pulse resulting therefrom arrives in these elements after different transit times - corresponding to the different position coordinates of the elements 1. From this, they generate electrical signals which are in the amplifiers 2 of each element 1 - possibly with preceding or subsequent analysis or filtering. amplified and then digitized in digitizers 3. The digitized signals derived from the converter signals or - with the interposition of an analysis or a filter - from their analysis values are fed to both the counter starter 9 and the amplification 4. The first incoming signal causes the counter starter 9 to start all the counters 5 of the receiver E assigned to it at the same time. The first signal reached the counter 5 via the delay element 4 only after the delay time and switched it off. The later arriving signals of the other channels of the receiver only reach their counter 5 after the delay time and additionally after the transit time difference. the channel in which the first signal arrived is. The other counters 5 are switched off correspondingly later, so that the differences in the counter values of the channels of a receiver correspond to the runtime differences. In parallel to the counting of the run-time differences, the absolute time at the start of the counter was measured by means of the time counter 7.

After all counters 5 of a receiver E have stopped, their values, controlled by the clock controller 8, are entered into the process computer 6, for which purpose the absolute time measured by the timer 7 when the counter 5 starts is stored as a parameter for the pulse. With a delay after the value for the absolute time has been accepted by the process computer 6, the counters 5 of the receiver E assigned to the computer 6 are deleted. Each computer 6 calculates, for the receiver E assigned to it, the transit time differences between the impulse and / or feature hit in the individual Wandlei elements 1 and the resulting direction from which the impulse has arrived at the receiver E.

The directional values of the micro-processors 6 of all the receivers E are entered into the computer 12 of the device, into which the coordinates of the receivers E have already been entered, which the computer 12 now relates to the directional values in a subsequent calculation step, after which it Defines the intersection of the directional straight line as the location of the sound source by superimposing the directions.

In the above-described analysis of the individual elements before digitization, the arrival of a certain characteristic of the sound pulse is determined measured. Instead, an analysis (eg frequency analysis) can take place parallel to the direction determination, which is directly correlated with the direction.

The directions of the receivers can also be saved and subsequently correlated with each other, which enables the processing of signals with very fast sequences. While in the known method, consequences of approx. 100 / sec. (depending on the distance between the recipients) could be processed, the processing is now independent of this distance and can have maximum values of approx. 1 - 2 million / sec. to reach. Other significant advantages over the prior art are that

the determination of the speed of the pulse from the source to the receiver becomes independent because the direction of the sound pulse is determined within the individual elements l:

- the impulses within the individual! emente l (distance of the elements <sound field diameter) remain similar, whereby the type of wave preferred by the respective receiver is unimportant and the resulting error is eliminated;

- Because the direction of the pulse is determined for each receiver, there is no longer any need to wait until the pulse has reached all the receivers required for location. The values of the individual recipients can be saved. since the impulses within the individual elements 1 remain similar, an analysis can be carried out easily in real time, which is directly correlated with the direction of the impulse, which is also determined immediately, or the characteristics of which determine the direction. In this way, the impulse can be clearly identified without having to take into account the usual transit time differences;

since the direction is immediately determined for each recipient, even complex geometries can be checked. By determining the direction, reflections and mode conversions can be easily taken into account and do not falsify the measurement result.

Claims

Process for sound emission location and analysis when testing equipment and components for damage, in which the position of the sound emission source is determined from the arrival of a sound pulse at various measuring points, characterized in that in three or more Adjacent sections (1) of each measuring point (E), the width of which is smaller than the sound field diameter to be expected, the time of arrival of a pulse or a pulse characteristic measured, from this the transit time difference for the sections (1) of each measuring point (E) determined and from these transit time differences for each measuring point (E) the sound arrival direction is determined, after which the position of the sound emission source is determined from the directional values and the position coordinates of the measuring points (E).

2. The method according to claim 1, characterized in that a sound correlated directly with the direction analysis. e.g. a frequency analysis is carried out at the individual measuring points (E).

3. The method according to claim l, characterized by an analysis or filtering of the signals of the individual sections (1) before their digitization.

4. Device for sound emission location and analysis during the testing of facilities and components for damage with multiple, to be attached at intervals on the surface of the test specimen the receiver probes (E), their transducers (1) convert the received sound pulses into electrical signals, as well as with the converters (1) downstream amplifiers (2), digitizers (3) and counters (5) and with a process computer (12) taking the counter values via a clock control (8) for determining the transit time difference between the sound pulses received by the individual transducers and for correlating them with the given receiver coordinates for the purpose of locating the sound emission source, characterized in that each receiver (E) has three or more separate transducer elements (1), whose overall width is less than the expected sound field diameter.

5. The device according to claim 4, characterized by each transducer element (1) downstream amplifier (2), digitizer (3) and counter (S) and a computer (microprocessor) (6) for each receiver (E), the takes over the counter values of the receiver (E) assigned to it and determines the direction in which the sound emission source, based on the respective receiver (E), is located, and by a further one connected downstream of the receiver computers (6), from the transit time differences obtained in this way Computer (12) for the location of the sound emission source from the directional values entered into it, the receiver computer (6) and the receiver coordinates by determining the directional intersection.

6. Device according to claims 4 and / or 5, ge. characterized by one of the digitizers (3) downstream counter starter (9) for simultaneous start of all counters (5) of a receiver (E) when the first pulse arrives.

7. The device according to claim 6, characterized by a delay element (4) upstream of each counter (5) with a delay time that is the same for all channels of a receiver (E).

8. The device according to at least one of the preceding claims 4 to 7, characterized by a timer (7) for all receivers (E) for measuring the time of the counter starts, the values of which are input to the receiver computer (6) via the clock control (8) .

9. The device according to at least one of the preceding claims, characterized in that each digitizer (3) is preceded by a filter or an analysis which only allows the characteristic feature to the counter (5).